Understanding World Happiness: Final Report
What is the significance of happiness in our current society and what factors contribute to making us happy? How can we measure happiness and better understand it? Happiness is a complex phenomenon to study as it is personal and subjective. What makes one individual happy might not be as significant to another. Since 2012, the Gallup World Poll has surveyed citizens of more than 137 countries across the globe on their life satisfaction to understand happiness and its determinants. By collecting data on people’s ratings on life satisfaction, Gallup aims to understand happiness and its determinants worldwide.
Happiness scores can provide important insights that can be used to inform policy. Since 2015, the highest happiness score recorded was 7.8/10, from Finland in 2021. Based on the Gallup World Poll data, Finland has important characteristics that contribute to a high level of happiness which include high GDP per capita, health/life expectancy, perceived familial/social support, freedom, and low perceived government corruption. Contrastingly, Afghanistan reported the lowest happiness score since 2015 and this was 1.7/10 in 2024. Afghanistan records low GDP per capita, health/life expectancy, and freedom. These factors potentially explain the low happiness score that citizens reported.
Comparing countries that vary greatly in their happiness can provide insight into which key factors contribute to happiness. However, happiness is complex, some countries do not fall on these extremes. Close analysis into country-level happiness distinctions can provide more useful insights as to which factors contribute the most to differing perceptions of happiness. This information is important for determining policy areas where more resources should be directed to improve people’s livelihoods. This project aims to unravel the most important factors contributing to happiness within countries and to understand why there is variation in happiness around the world.
We use data provided by the Gallup World Poll from 2015 to 2024 which contains national average responses to questions of life evaluations. There are between 137 and 159 countries represented in the survey, depending on the year. There are seven quantitative variables in the survey, including; happiness score, GDP per capita, perceived social/familial support, healthy life expectancy, perceptions of freedom, feelings of generosity, and perceptions of government corruption. Measures of each variable are based on average calculations which will be further described within the Checks section. Furthermore, an in-depth description of each variable is found within the Appendix.
Initial Steps
In its original form, there were 10 different data files representative of each year of data collections from 2015 to 2024. Our first step was to learn about the number of variables, the particular countries, and the different variables in each of the datasets and to confirm that the information was accurate and consistent between datasets. We learned of inconsistencies within the yearly data. There were some years that had more countries than others and we had to note this. There were also a few variables that were coded differently in some datasets. For example, in earlier years, what we standardized as a perceived family/social support variable, was described as ‘Social support’ in some years. To this, changes were made to standardize this data, as we worked to recognize which columns were the same though under a different name. And, which variables to not include within the data as these were not measured consistently throughout the years. To fix missing region data, we completed a join with another data set which contained region and country information to ensure that all of the countries’ for all years also had their appropriate region information. In the compilation of the data sets and pre-processing steps we received assistance from Professor Heggeseth. This was a part of the data cleaning and pre-processing steps.
The full combined data set consists of 10 columns and 1,510 rows of data. Each row is an observation containing the collected data on one country’s happiness report for one year. The columns present are country, happiness score, economic GDP per capita, perceived familial/social support, healthy life expectancy, perceptions of freedom, perceptions of government corruption, feelings of generosity, year, and region of the world. The structure of the data consists of two qualitative variables and eight quantitative variables. The qualitative variables describe country and region. The eight qualitative variables provide information about the people within the countries. Within checking the data the NA count stood out, to this we got rid of any missing values from the columns.
Based on Figure 1, we can infer that the distribution of happiness since 2015 has changed. There is a slight rightward shift in happiness score distribution suggesting that a number of countries are getting happier in general over the last decade. Many countries have moved from having happiness scores in the middle range of 3.5 to 6.5 to slightly higher tier in the 5.5 - 7.0 range. This increase is cause to dive deeper into investigating whether all countries alike are experiencing this increase in happiness, or if this is unique to certain countries, and then to further understand, what factors contribute to this net increase in happiness.
Figure 2. Mean Happiness Score by Country in 2024
To answer the question of whether this change is consistent geographically, we show the changes in happiness levels in Figures 2 and 3. Generally, there is a slight increase in happiness levels in some regions of the world. In the Americas, we can observe a few countries that have increased in their happiness levels such as Brazil, the United States and Canada. In Asia, countries such as China also see a slight increase in happiness levels. This confirms the trend observed in Figure 2 where countries are on an upward trajectory. We can also note that many countries somewhat remain at around the same levels of happiness in 2024 that they were in 2015. While the distribution of happiness has shifted positively, there are also many countries that remain at the same levels of happiness, whether that is a low, average or high score. Additionally, in both 2015 and 2024, we can observe that countries vary widely in their happiness levels. In general, countries in the Americas, Europe, as well as Australia have higher levels of happiness compared to countries in Asia and Africa.
Next we consider how happiness varies by region, exploring which individual countries are contributing to this overall increase in global happiness and which countries are not. Above we can see a graph showcasing the change in happiness per world region from 2015 to 2024, highlighting countries with prominent changes in happiness. Countries that have over a +/- 1.1 unit change in the level of happiness, are colored dependent on if that change is positive or negative. Changes greater than 1.1 are considered “positive.” These are colored in green. Changes less than 0 we consider “negative” and color in dark purple. All changes between a 0 to 1.1 unit change in the level of happiness we consider “stable” and color in grey.
The majority of countries with drastic changes had positive changes. Though all countries in East Asia, Southeast Asia and North America and ANZ regions, had all grey lines, meaning they had experienced stable changes in happiness score. Prominent decreases in happiness are found in Lebanon, Afghanistan, Zambia, Zimbabwe, Sierra Leone and Venezuela (respectively). Prominent increases in happiness are found in Ivory Coast, Togo, Guinea, Romania, Serbia, Bulgaria, Hungary, Gabon, Honduras, Latvia, Armenia (respectively). The question we can propose after exploring this relationship is, what are the factors contributing to a country’s drastic change in happiness?
To investigate factors that contribute to dramatic changes in happiness, we plot a stacked bar chart that breaks down factors that contribute to happiness in a select number of countries. This graph takes into account the countries that have been previously distinguished in Figure 4, to have more dramatic increases in happiness from 2015 to 2024. The countries include Latvia, Hungary, Armenia, Venezuela, Honduras, Lebanon, Afghanistan, Zimbabwe, Zambia, Sierra Leone, Ivory Coast and Guinea. Then, this graph displays the different factors contributing to each of these country’s happiness score as their contribution to the countries’ total happiness score.
Within the graph, where the 100% is marked, the blank space between 0 and 100% is representative of the percent unexplained by the factors. This graph then animates over the course of the years from 2015 to 2024 where the viewer is able to observe the increase/decrease of each marked factor for each county over the years.
Overall, we find that family, economic_gdp_per_capita, and life_expectancy have the biggest impacts on a country’s overall happiness score. We also observe that a person’s perceptions of government corruption does not have a very big impact on a country’s overall happiness score.
Based on our analysis, we have observed that the happiest regions are North America and ANZ and Western Europe. Comparing the different factors that contribute to happiness, we find that GDP per capita, health/life expectancy, and social support have the greatest impact on a country’s happiness score. This suggests that people’s immediate physical needs play a greater role in their happiness. While other factors do not seem as significant, it is worth mentioning that they also have a role to play, although smaller.
Some further questions that can be explored by future research are, How do cultural differences help explain people’s responses and their perception of happiness, and how can we further explore the unobserved factors?
While this study provides valuable insights, it is important to acknowledge its limitations. Happiness scores are approximated using external measures such as social, economic, and political structures, which can have a limiting scope.
Furthermore, within data collection, Gallup World Poll utilizes telephone surveys in countries where telephone coverage represents >80% of the population. In the developing world, Gallup uses an “area frame design”— face-to-face interviewing in random households. It is important to consider: all samples are based on ease of access and nationally representative of the resident population 15 and older, with some exceptions. The coverage area is the entire country, and the sampling frame: the entire civilian, non-institutionalized adult population. Exceptions include areas where safety of the interviewing Gallup staff was threatened and scarcely populated islands within some countries.
The typical survey includes >1,000 individuals. In some large countries such as China and Russia, sample sizes of at least 2,000 are collected while in rare cases, the sample size is between 500 and 1,000. In this, we must consider who is excluded by these data collection methods. We cannot consider a countries’ data as comprehensive for the countries’ overall opinion. People choosing not to pick up the phone might have different opinions. Those skeptical of government corruption might not answer honestly, or choose not to answer the questions at all. Incarcerated citizens weren’t given surveys.
It is important to understand that Gallup weights World Poll samples to correct for unequal selection probability, non-response, and double coverage of land-line and cellphone users when using both cellphone and land-line frames. Gallup also weights its final samples to match the national demographics of each selected country. In addition to sampling error, question wording and practical difficulties in conducting surveys can introduce error or bias into the findings of public opinion polls.
Additionally, it is important to consider the fact that this data is not collected longitudinally, but cross-sectionally. Different opinions are received from different people every year. Comparing happiness levels between different people per year creates gaps based on which populations of people are being surveyed each year. In this, we must use the data and make claims conscientiously.
It is important to understand that Gallup weights World Poll samples to correct for unequal selection probability, non-response, double coverage of land-line and cellphone users, and to match the demographics of each selected country. In addition sampling errors, question wording and practical difficulties in conducting surveys can introduce error/bias.
As data is collected, used and interpreted, it is important to keep these factors in mind.
Gallup, Inc. (2014, October 14). How does the Gallup World Poll work? Retrieved December 5, 2024, from Gallup.com website: https://www.gallup.com/178667/gallup-world-poll-work.aspx
Helliwell, John F., Richard Layard, and Jeffrey Sachs, eds. 2015. World Happiness Report 2015. New York: Sustainable Development Solutions Network.
Helliwell, J., Layard, R., & Sachs, J. (2016). World Happiness Report 2016, Update (Vol. I). New York: Sustainable Development Solutions Network.
Helliwell, J., Layard, R., & Sachs, J. (2017). World Happiness Report 2017, New York: Sustainable Development Solutions Network.
Helliwell, J., Layard, R., & Sachs, J. (2018). World Happiness Report 2018, New York: Sustainable Development Solutions Network.
Helliwell, J., Layard, R., & Sachs, J. (2019). World Happiness Report 2019, New York: Sustainable Development Solutions Network.
Helliwell, John F., Richard Layard, Jeffrey Sachs, and Jan-Emmanuel De Neve, eds. 2020. World Happiness Report 2020. New York: Sustainable Development Solutions Network
Helliwell, John F., Richard Layard, Jeffrey Sachs, and Jan-Emmanuel De Neve, eds. 2021. World Happiness Report 2021. New York: Sustainable Development Solutions Network.
Helliwell, J. F., Layard, R., Sachs, J. D., De Neve, J.-E., Aknin, L. B., & Wang, S. (Eds.). (2022). World Happiness Report 2022. New York: Sustainable Development Solutions Network.
Helliwell, J. F., Layard, R., Sachs, J. D., Aknin, L. B., De Neve, J.-E., & Wang, S. (Eds.). (2023). World Happiness Report 2023 (11th ed.). Sustainable Development Solutions Network.
Helliwell, J. F., Layard, R., Sachs, J. D., De Neve, J.-E., Aknin, L. B., & Wang, S. (Eds.). (2024). World Happiness Report 2024. University of Oxford: Wellbeing Research Centre.
Sustainable Development Solutions Network. (2019). World Happiness Report [Data set].- Happiness Score - (Sometimes referred to as the Happiness ladder.) The national average response to the question (Please imagine a ladder, with steps numbered from 0 at the bottom to 10 at the top. On which step of the ladder would you say you personally feel you stand at this time?”) of life evaluations.
- GDP per capita - Measures purchasing power parity (PPP) at constant 2011 international dollar prices are from the November 6, 2014 release of the World Development Indicators (WDI).
- Social Support / Family - Average of the binary responses (either 0 or 1) to the GWP question “If you were in trouble, do you have relatives or friends you can count on to help you whenever you need them, or not?”
- Health/Life Expectancy - Healthy life expectancies at birth are based on the data extracted from the World Health Organization’s (WHO) Global Health Observatory data repository. They use a ratio of the “Healthy life expectancy” : “Total life expectancy.” Therefore, the higher the ratio, the better.
- Freedom - National average of responses to the question “Are you satisfied or dissatisfied with your freedom to choose what you do with your life?”
- Generosity - Residual of regressing national average of responses to the question “Have you donated money to a charity in the past month?” on GDP per capita.
- Corruption / Government - This variable measures the national average responses to the two questions “Is corruption widespread throughout the government or not?” and “Is corruption widespread within businesses or not?” The overall perception is just the average of the two 0-or-1 responses.